Item-Level Domain Analysis May Interpret Quality of Life Responses in SLE

Item-level analysis provides clinicians with a new avenue for interpreting health-related quality of life (QoL) response variations among patients with systemic lupus erythematosus (SLE), according to research results published in Rheumatology. This analysis type may be easily implemented in future clinical trials.

Using data from the French Longitudinal Study of Quality of Life and Activity in Systemic Lupus Erythematosus (EQUAL) study (ClinicalTrials.gov Identifier: NCT01904812), researchers sought to explore the effect of disease activity on specific health-related QoL item responses in a population of patients with SLE. Data from the longitudinal, multicenter cohort study were collected between December 2011 and July 2015.

In total, 336 patients were included in the cohort (302 [89.9%] women; mean age, 41.2±11.9 years). Duration of SLE was >10 years for 147 patients (43.8%); 131 (39.0%) had a Systemic Lupus International Collaborating Clinics/American College of Rheumatology damage index ≥1.

At baseline, the median Safety of Estrogens in Lupus Erythematosus National Assessment (SELENA)-SLE Disease Activity Index within the cohort was 2 (interquartile range, 6). In terms of health-related QoL, mean scores varied by domain; the mean score for the emotional health domain was 71.3±23.7 (measured by LupusQoL), while the physical health domain was 85.6±19.9 (measured by SLEQOL). Among patients who experienced flares, the most common were musculoskeletal flares (28%), followed by cutaneous flares (15.5%); other types of flares included neurologic, psychiatric and renal. Overall, most flares were mild or moderate.

Study results were broken down by flare type as follows:

Musculoskeletal flares: For LupusQoL items, patients in the remitting group with musculoskeletal flares significantly improved in items 7 and 10 of the physical health and pain domains (“doing things slower” and “pain interferes with the quality of sleep”) at 6- and 18-month follow-up visits. However, responses to item 30 of the body image domain (“feeling less attractive due to weight gain”) were significantly affected 6 months after a flare; health-related QoL worsened in patients who experienced flares compared with patients who did not experience flares. For SLEQOL items, significant improvements were noted in item 6 of the physical functioning domain (“difficulty for walking 3 km”), items 22 and 23 of the symptoms domain (“troubled by sore skin” and “troubled by joint pain”), and items 24 and 27 of the treatment domain (“troubled by fear of needles” and “troubled by inconvenience of clinical visits”).

Cutaneous flares: For LupusQoL items, patients in the remitting group experienced significantly better results compared with those who did not flare in terms of the pain domain at each follow-up visit, except for item 10 at 18 months. Significant improvements were noted in responses for item 23 of the emotional health domain (“feeling anxious”). Improvements occurred at 6 months for items 15 and at 6 and 18 months for item 16 (“less interested in a sexual relationship because of pain” and “not interested in sex,” respectively). For SLEQOL items, patients in the remitting group experienced significantly greater positive change compared with patients who did not flare in items 1 through 4 of the physical functioning domain (“difficulty for walking outdoors, shopping, turning taps, going to the market”) at 18 months and at all time points for item 6 (“difficulty for walking 3 km”). Improvements were also noted in items 17, 18, 22, and 23 of the symptoms domain (“troubled by loss of appetite,” “troubled by fatigue,” troubled by sore skin,” and “troubled by joint pain”), and significant improvements were noted for item 40 of the self-image domain (“troubled regarding alcohol or tobacco consumption”) at months 6 and 18.

Renal flares: For LupusQoL items, significant improvements were seen in patients in the remitting group for item 2 of the physical health domain (“help for moderate physical jobs”) and at 6 and 18 months for item 14 in the planning domain (“difficulty to commit to social arrangements”). Patients in this group also demonstrated significant improvement in item 29 of the body image domain (“feeling less attractive due to hair loss”). For SLEQOL items, a significantly greater increase in health-related QoL was seen in patients in the remitting group for item 4 of the physical functioning domain (“difficult going to the market”). Patients in the remitting group also experienced improvements across the activities domain at 6 and 18 months. Compared with patients who did not experience a flare, there were significant improvements in item 17 of the symptoms domain (“troubled by loss of appetite”) 18 months after the renal flare.

Constitutional flares: For LupusQoL items, significant improvement was noted in item 3 of the physical health domain (“help for light physical jobs”) at months 6 and 12. Similarly, items across the ‘burden to others’ domain improved significantly after 6 or 12 months. For constitutional flares, a 6-month improvement was noted in items 1, 2, and 4 of the physical functioning domain (“difficulty walking outdoors, shopping, turning taps, and going to the market” and item 6 of the activities domain (“troubled regarding sex-related activities”). Patients in the remitting group worsened in item 25 of the treatment domain (“troubled by dietary restrictions”).

Neurologic flares: For LupusQoL items among patients in the remitting group, health-related QoL significantly improved in item 33 of the fatigue domain (“need to have early nights”) at both 6 and 18 months. However, item 27 of the body image domain worsened significantly (“avoiding social situations because of appearance”). For SLEQOL items, significantly improved responses were noted for items 29 through 31 of the mood domain (“troubled by feeling low, depression, anxiety”) at months 6 and 12.

Study limitations included the grouping of mild, moderate, and severe flares into a single study because of the low occurrence of moderate and severe flares within the cohort, possibly leading to an underestimation of the effect of specific flare types on QoL items. In addition, the role of ethnicity could not be explored in depth because of limitations by French regulatory authorities.

“Item-level analysis provides a new way of interpreting [health-related] QoL variation in [patients with] SLE,” the researchers concluded, “thus permitting (i) better understanding of disease activity impact…and (ii) [providing] a more personalized follow-up of patients according to the expected [effect] of a given clinical phenotype on [health-related] QoL items.”

Disclosure: Several study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.

Reference

Corneloup M, Maurier F, Wahl D, et al; for the EQUAL study group. Disease-specific quality of life following a flare in systemic lupus erythematosus: an item response theory analysis of the French EQUAL cohort [published online October 17, 2019]. Rheumatology. doi:10.1093/rheumatology/kez451